An Elman Neural Network Application on Dynamic Equivalents of Power System

This paper presents an Elman neural network based on Genetic algorithms for the identification of dynamic equivalents of power system. The Elman neural network is one of the dynamic recurrent neural networks. In this paper, a modified Elman network is introduced first. Then we propose its training algorithm using Genetic algorithms. Lastly, the proposed method is demonstrated and compared with the original system using the 9 machines 36 buses EPRI test system. Simulation results show that the Elman network based on GAs can achieve favorable effects on the application of dynamic equivalents of power system.